Machine learning techniques for implicit interaction using mobile sensors

Interactions in mobile devices normally happen in an explicit manner, which means that they are initiated by the users. Yet, users are typically unaware that they also interact implicitly with their devices. For instance, our hand pose changes naturally when we type text messages. Whilst the touchscreen captures finger touches, hand movements during this interaction however are unused. If this implicit hand movement is observed, it can be used as additional information to support or to enhance the users’ text entry experience. This thesis investigates how implicit sensing can be used to improve existing, standard interaction technique qualities. In particular, this thesis looks into enhancing front-of-device interaction through back-of-device and hand movement implicit sensing. We propose the investigation through machine learning techniques. We look into problems on how sensor data via implicit sensing can be used to predict a certain aspect of an interaction. For instance, one of the questions that this thesis attempts to answer is whether hand movement during a touch targeting task correlates with the touch position. This is a complex relationship to understand but can be best explained through machine learning. Using machine learning as a tool, such correlation can be measured, quantified, understood and used to make predictions on future touch position. Furthermore, this thesis also evaluates the predictive power of the sensor data. We show this through a number of studies. In Chapter 5 we show that probabilistic modelling of sensor inputs and recorded touch locations can be used to predict the general area of future touches on touchscreen. In Chapter 7, using SVM classifiers, we show that data from implicit sensing from general mobile interactions is user-specific. This can be used to identify users implicitly. In Chapter 6, we also show that touch interaction errors can be detected from sensor data. In our experiment, we show that there are sufficient distinguishable patterns between normal interaction signals and signals that are strongly correlated with interaction error. In all studies, we show that performance gain can be achieved by combining sensor inputs.

Synthesis of novel organic semiconductors for optoelectronic devices

This thesis describes the synthesis and characterisation of novel conjugated organic materials with optoelectronic application. The first chapter provides an introduction about organic semiconductors and in particular about their working principle from a physical and chemical point of view. An overview of the most common types of solar cells is provided, including examples of some of the best performing materials. The second chapter describes the synthesis of a new library of flavin derivatives as potential active materials for optoelectronic applications. Flavins are natural redox-active molecules, which show potential application in optoelectronics, thanks to their stability and versatility. FPF-Flavins, for instance, could be used either as acceptor units in push-pull polyconjugated systems or as acceptor unit in dyes for DSSCs. In the same chapter a first attempt of synthesising bis-flavins to be used as N-type semiconductors in BHJ devices is described. The third chapter describes the successful synthesis and characterization of a series of conjugated organic molecules based on the benzothiadiazole moiety. Among these, three molecules containing ferrocene as donor unit were tested as sensitizers for DSSCs, reporting a PCE of 0.3% as the best result. Further studies indicated a significant problem of charge recombination which limits the performance. A near-infrared absorbing push-pull polymer, based on BbT as acceptor unit, was also synthesised and tested in BHJ devices as P-type semiconductor in blend with PC71BM, showing a VOC of 0.71 V. Finally, the last chapter describes the synthesis of several tetrathiafulvalene derivatives in order to explore this moiety as donor unit in dyes for DSSCs and as HTM for perovskite-based solar cells. In particular, two very simple dyes were synthesised and implemented in DSSCs reporting a PCE 0.2% and 0.4%, respectively. The low efficiency was associated to the tendency to aggregate at the solid state, with the absorption shifting from the visible to the infrared range. A conjugated molecule, containing a DPP core, was also synthesised and tested as HTM for perovskite solar cells. The best reported PCE of 7.7% was obtained without any additives. A case study about dehalogenation and “halogen dance” in TTF iodide is also presented.